
The author introduces MAGNET, a masked generative sequence modeling method for audio generation using a single non-autoregressive transformer. The approach combines efficient training with a novel rescoring method to enhance the quality of generated audio.


coremsg

Masked-Audio-Generation-using-a-Single-Non-Autoregressive-Transformer


Masked Audio Generation using a Single Non-Autoregressive Transformer


title_rewrite


The author aims to improve audio generation diversity within specific categories by incorporating visual information, utilizing a clustering-based method. This approach enhances the quality and diversity of generated audios significantly.


Enhancing-Audio-Generation-Diversity-with-Visual-Information


Enhancing Audio Generation Diversity with Visual Information
